JP2006094491A - Method and apparatus for processing a video signal to compensate for defects in a display device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for reducing a predetermined visual defect by applying different gamma rows in each of horizontal and vertical processing directions.
The present invention relates to a method and apparatus for processing a video image for the purpose of correcting the result of a display device. According to the invention, the method comprises the following steps: Correction of the video signal by a specific gamma law associated with the first component of the horizontal and vertical components of the video signal. A first video process that operates on a first component of the corrected video signal. Application of intermediate rows to the processed video signal that affect transmission from a space that is linear with respect to the first component to a space that is linear with respect to the other of the horizontal and vertical components called the second component. A second video processing that acts on the second component of the video signal emitting from the previous step. Correction of the video signal processed by the specific inverse gamma law associated with the second component.
[Selection] Figure 3

Description

  The present invention relates to a method and apparatus for processing video images, particularly for the purpose of compensating for defects in certain display devices such as non-linear display devices (eg cathode ray tubes).

  A video signal may be represented in various spaces connected by a transfer function. Thus, the color level representation space consists of a set of integers that vary between 0 and 255, for example. The visualization space is a space where a light signal is generated and perceived for a part thereof.

  In most video image display devices, the color level representation space and the visualization space are related by a non-linear law called electro-light transfer function or more generally by gamma law. The emission intensity reproduced on the screen is a non-linear function of the color level at the input of the display device.

  The transfer function associated with the two representation spaces is a characteristic of the type of display device. In the case of a cathode ray tube, the gamma low of the cathode ray tube has the following formula:

ｎ_inは入力レベル、ｎ_outは出力レベル、Ｎ_inはｎ_inを特徴付ける最大レベル、Ｎ_outはｎ_outを特徴付ける最大レベル、γは伝達関数を特徴付ける係数である。

n _in is an input level, n _out is an output level, N _in is a maximum level characterizing n _in , N _out is a maximum level characterizing n _out , and γ is a coefficient characterizing a transfer function.

  According to the prior art, in order to estimate the gamma low of a display device, it is known to display on the display device a series of uniform images whose levels vary from 0 to 255. The light level L (n) emitted by the screen is then measured for each input level n and the curve is normalized with the maximum of these input levels. γ = 255 * L (n) / L (255).

  FIGS. 1 and 2 illustrate a prior art processing method for video signals that have been pre-corrected (FIG. 1) or not pre-corrected (FIG. 2) with reverse gamma low for the purpose of reducing visual artifacts by the display device. is doing.

  The transfer function described above and referred to by reference numeral 15 in FIG. 1 (hereinafter also referred to as “posterial”) has the property of attenuating dark pixels. It is known to pre-correct the initial video signal originating from the camera with an inverse gamma low 11 (hereinafter also referred to as an ante-low) so as to correct this attenuation before sending the correction signal to the display. . This process commonly performed at the camera level is not compulsory. The gamma value used to pre-correct the initial video signal is known and defined in various literature, especially in the ITU recommendations of television (eg ITU-R BT.709-5, Basic Parameters Value for HDTV, ITU-R BT.470-6, Conventional Television System).

  Further, prior to displaying the transmitted video signal on the screen, for example, video processing 13 is applied to the pre-corrected or non-pre-corrected video signal so as to improve the quality of the displayed video signal. It may be necessary to do. Video processing may be, for example, re-interpolation of a video image for the purpose of correcting distortion in the image. The distortion is due to the display device.

  When such video processing is applied, it is necessary to work in a space that is linear with respect to the visualization space. In particular, in the opposite case, the state of the defect dependent on the processing is seen in the displayed image. For example, in the case of re-interpolation, the brightness of an image point may vary depending on whether it has been re-interpolated for one pixel or n pixels. In particular, in this case, the visual sum of n pixel intensities is not the same as the visual sum of the pixels alone if the processing space is not linear with respect to the visualization space.

  Thus, as illustrated in FIG. 1, in blocks 12 and 14, respectively, on either side of processing block 13, if the first signal has been pre-corrected, the anterior row 11 is inverted and the posterial row 15 is inverted. It is known to do. The display device makes corrections according to the gamma-low characteristics of the display device, thereby correcting the video signal.

  However, as previously described, the video signal may be generated and generated by the camera 11 without being pre-corrected beforehand by the inverse gamma law 11. In this case, it is known to apply video processing directly to the initial video signal 21 as illustrated in FIG. The signal thus obtained is corrected by the inverse gamma law 22 so as to correct the dark point attenuation introduced by the transfer function of the cathode ray tube. The display device 23 corrects the video signal thus corrected in accordance with the gamma-low characteristics of the display device.

  Prior art methods can correct certain defects (eg, dark point attenuation), but depend on the direction of visualization (eg, overlap between pixels, spot shape). The predetermined characteristic of the display device is not taken into consideration.

  Thus, in the case of application of video processing applied in succession to the horizontal component (X) and vertical component (Y) of the image, the application of global gamma low may cause visual defects. The term “global gamma law” means a law that applies in the same way whether to process the vertical or horizontal component of a video signal. Local attenuation defects in the red, green and blue signals that cause color defects can be seen, for example, when applying a color image re-interpolation process.

  It is therefore desirable to consider the (horizontal or vertical) direction of processing in order to process the video signal.

  To that end, the present invention provides a method whose purpose is to reduce certain visual defects by applying different gamma rows in each of the (horizontal and vertical) processing directions. According to the present invention, it is proposed to process a video signal including a horizontal component and a vertical component before the signal is displayed by the display device, and the following steps are followed.

Correction of the video signal by a specific gamma law associated with the first of the horizontal and vertical components of the video signal.
A first video process that operates on a first component of the corrected video signal.
Application of intermediate rows to the processed video signal that affect transmission from a space that is linear with respect to the first component to a space that is linear with respect to the other of the horizontal and vertical components called the second component.
Second video processing that acts on the second component of the video signal output from the previous step.
Correction of the video signal processed by the specific inverse gamma law associated with the second component.
The order of processing of the signal components is not important.

  According to a preferred embodiment, the intermediate gamma law is the wax associated with the second component and the inverse gamma wax component associated with the first component. In particular, after processing the signal according to one of these components, it is appropriate to take a standpoint in a space that is linear according to the other component. Therefore, when the Y component is processed first, it is appropriate to incorporate the viewpoint in a linear space in X.

  According to a preferred embodiment, the video signal is corrected by various gamma rows, and the intermediate row uses a correspondence table that associates the output level calculated according to the appropriate row with the respective input level of the video signal. Applied.

Advantageously, the first and second video processing is a process of re-interpolation of the video signal according to the first component of the signal and according to the second component of the signal.
According to a preferred embodiment, the reinterpolation of the video signal consists of signal filtering and the filter used is a polyphase filter.

According to a preferred embodiment, the resolution in terms of the number of bits (or dynamic range) of the video signal is increased during a correction step with a specific gamma law associated with the first component of the video signal, During the correction step with the specific inverse gamma law associated with the second component of the signal, the resolution is first returned to resolution. The increase in resolution in terms of the number of bits can eliminate the remaining quantization artifacts.
The present invention is advantageously applied to non-matrix display devices.

The invention also relates to an apparatus intended to implement the method. According to the invention, the device comprises:
Means for correcting the video signal by a specific gamma law associated with the first component of the horizontal and vertical components of the video signal.
Means for applying a first video processing acting on the first component of the corrected video signal;
Apply an intermediate row to the processed video signal that affects transmission from a space that is linear with respect to the first component to a space that is linear with respect to the other of the horizontal and vertical components, referred to as the second component, For generating a simple video signal.
Means for applying a second video processing acting on the second component of the intermediate video signal;
Means for correcting the video signal processed by a particular inverse gamma law associated with the second component of the video signal.

  The invention will be understood and illustrated by means of advantageous exemplary modes, which are not restrictive as a whole of embodiments and implementations, with reference to the accompanying drawings.

  The method according to the invention is illustrated by FIGS. In FIG. 3 and FIG. The method includes five steps indicated by steps 32-36 in FIG. 3 and five steps indicated by steps 41-45 in FIG.

The first step 32 of method 3 according to the invention (step 41 in method 4) consists in using Y in linear space. Therefore, the vertical row γ _Y characteristic is applied to the first video signal that has been pre-corrected 31 by a global inverse gamma row. To apply this row, it is known to use a correspondence table that associates the output level calculated according to the appropriate inverse gamma row with each input level.

  During this operation, the resolution is increased in terms of the number of bits. This in particular makes it possible to remove certain quantization artifacts. Thus, the subsequent steps can be performed with a larger dynamic range in terms of number of bits, and the return to the initial dynamic range is effected in response to the last gamma row application.

In the case where the initial video signal has been pre-corrected by the reverse global gamma law, the law γ _Y 32 is the composition of the law Ψ _Y and the global gamma law, ie γ _Y = Ψ _Y oγ. Where o is the method of composition of the function and Ψ _Y corresponds to the deviation between the global gamma law and the law γ _Y measured along Y. γ _Y is estimated directly according to the procedure described later in this specification. The gamma low is known and the low Ψ _Y is estimated from this.

In the case where the first video signal has not been pre-corrected by the reverse global gamma row, row γ _Y 41 is equal to row ψ _Y , ie γ _Y = ψ _Y.

  In order to correct the distortions introduced by the display device, video processing is applied to the video signal which makes it possible to correct these distortions. This process may be, for example, filtering of video sequence images. More precisely, it is possible to apply an image re-interpolation process. It is also possible to apply another video processing that addresses the specific technical requirements of the display device.

  For this reason, the second step 33, 42 according to the invention consists of reinterpolating the video image with Y by applying to the pixel a filter that depends on the displacement of Y to be acted upon. For example, it is possible to apply a polyphase filter whose phase is calculated based on the shift in the position of the point where it is to be calculated. Thus, the phase is zero when the point is in the first grid. If the point is shifted by 1/2 of the pixel with respect to the first grid, the phase will be +0.5. Other types of filters that can re-interpolate the image can be applied (eg, bicubic, bilinear filters).

According to the present invention, after processing the vertical component of the video signal, the horizontal component of the video signal is processed. It is then necessary to use stand points in linear space with X. For this reason, in the third steps 34, 43, the following intermediate function γ _intermediate = γ _X oγ _Y ⁻¹ is applied to the video signal resulting from the previous step. For this step, it is also possible to use a correspondence table that associates the output levels calculated according to the row γ _intermediate with the respective input levels. The function γ _X is estimated by the procedure described later in the description.

According to the invention, the fourth step 35, 44 consists of re-interpolating the video image at X by applying to the pixels a filter that depends on the displacement of X to be applied.
Finally, the fifth step 36, 45 consists of correcting the processed signal by applying the horizontal row γ ^-1 _X characteristic to the signal. For this step, it is also possible to use a correspondence table that restores the initial dynamic range if changed.

It is also possible to start by processing the signal along the horizontal component X during the five steps described above and then continue processing the signal along the vertical component Y.
The video signal processed in this way is then displayed on the display devices 37 and 46.

According to the present invention, the procedure described above for estimating gamma low according to the prior art is adapted to estimate the rows γ _X and γ _Y. Therefore, in order to estimate γ _X , a horizontal line that is one pixel higher, separated by a sufficient number of black lines, is displayed, and the influence of Y that may intervene from pixel to pixel is eliminated (eg, 4 black line). The light level L (n) emitted by the screen is then measured for each input level n and the curve is normalized by the maximum of these input levels. γ _X = 255 * L (n) / L (255). In the same way, proceed to estimate γ _Y using a 1 pixel wide vertical line separated by a sufficient number of black columns.

The method of the present invention can also be applied to rows other than gamma rows for correcting video signals.
The invention also relates to a device 5 linked to a display device 54 intended to implement the method described above. Only the essential elements of the device are shown in FIG. The device according to the invention in particular includes a memory 52, which stores correspondence tables used for processing by the various gamma rows 32, 34, 36, 41, 43, 45. The apparatus also has a processing unit 53 such as a microprocessor containing a processing program such as a re-interpolation program. The processing unit also has a conversion function that enables processing by various gamma rows 32, 34, 36, 41, 43, and 45. These functions use a correspondence table stored in the memory. In addition, the apparatus includes an input / output interface 51 for receiving an input video signal pre-corrected by gamma law and sending the video signal processed according to the methods 3 and 4 of the present invention to a display device.

FIG. 6 is a diagram illustrating a method of processing a video signal according to the related art for the purpose of reducing visual artifacts caused by a display device in a case where an initial video signal is pre-corrected by a reverse gamma law. FIG. 6 is a diagram illustrating a method of processing a video signal according to the related art for the purpose of reducing visual artifacts caused by a display device in a case where the initial video signal is not pre-corrected by a reverse gamma law. FIG. 6 is a diagram illustrating a method for processing a video signal according to the present invention in order to reduce visual artifacts caused by a display device in a case where an initial video signal is pre-corrected by a reverse gamma law. FIG. 6 illustrates a method for processing a video signal according to the present invention in order to reduce visual artifacts caused by a display device in a case where the initial video signal has not been pre-corrected by a reverse gamma law. FIG. 2 illustrates an apparatus for processing a video signal according to the present invention for the purpose of reducing visual artifacts caused by a display apparatus.

Explanation of symbols

31: Pre-correction 32, 34, 36, 41, 43, 45: Gamma low 33, 35, 42, 44: Video processing 37, 46: Display device

Claims (8)

A method of processing a video signal including a vertical component and a horizontal component before display of the video signal by a display device, comprising:
Correcting the video signal with a specific gamma law associated with the first component of the horizontal and vertical components of the video signal;
Applying a first video processing acting on a first component of the corrected video signal;
Applying an intermediate row to the processed video signal that affects transmission from a space that is linear with respect to the first component to a space that is linear with respect to the other of the horizontal and vertical components, referred to as a second component; ,
Applying a second video processing acting on a second component of the video signal output from the previous step;
Correcting the video signal processed by a particular inverse gamma law associated with a second component of the video signal;
A method characterized by comprising: The intermediate gamma wax is a composition comprising a wax associated with the second component and an inverse gamma wax associated with the first component.
The method of claim 1 wherein: The correction by the gamma row and the application of the intermediate row are performed by a correspondence table that associates an output level calculated according to an appropriate row with each input level of the video signal.
The method according to claim 1 or 2, characterized in that The first and second video processes are re-interpolation processes of the video signal according to the first component and the second component of the video signal, respectively.
4. A method according to any one of claims 1 to 3, characterized in that Re-interpolation of the video signal comprises filtering of the video signal, and the filter used is a polyphase filter,
The method of claim 4 wherein: The resolution in terms of the number of bits of the video signal is increased during a correction step with a specific gamma law associated with the first component of the video signal, and a specific inverse associated with the second component of the video signal. The first resolution in terms of the number of bits during the gamma-low correction step of
The method of claim 1 wherein: Applied to non-matrix display devices,
6. The method of claim 5, wherein: An apparatus for processing a video signal including a vertical component and a horizontal component before display of the video signal by a display device,
Means for correcting the video signal with a specific gamma law associated with a first of the horizontal and vertical components of the video signal;
Means for applying a first video processing acting on a first component of the corrected video signal;
Applying an intermediate row to the processed video signal that affects transmission from a space that is linear with respect to the first component to a space that is linear with respect to the other of the horizontal and vertical components, referred to as the second component, Means for generating a typical video signal;
Means for applying a second video processing acting on a second component of the intermediate video signal;
Means for correcting the video signal processed by a particular inverse gamma law associated with a second component of the video signal;
A device characterized by comprising:

Images